TNC Repositioning

TNC Repositioning#

There can be an uneven spatial distribution of TNCs that can be exacerbated over time with uneven incoming trip requests. Under zone-based vehicle-passenger assignment architecture, it is possible the truncated list of neighboring zones can restrict further assignment of TNC vehicles if these TAZs are positioned outside a free-flow minimum LOS threshold. Therefore, vehicles also need to be able to perform repositioning trips in addition to pickup and dropoff tasks to better balance the supply of TNC vehicles.

There are three TNC repositioning strategies:

default
joint_evcr
day_ahead_charging

Default#

The default repositioning strategy is applicable for non-electric vehicles. The strategy is decentralized, such that the TNC vehicles performing repositioning are based on their current location, and demand (# of requests), and supply (# of vehicles) aggregated at the TAZ level. This strategy uses the GLPK solver (which may require you to configure and build POLARIS again).

Joint-EVCR#

The joint_evcr strategy is an optimization-based dispatch strategy for repositioning and charging of SAEVs. The strategy is solved at discrete intervals in the simulation (e.g., quarter hours). This strategy uses the CPLEX solver (which may require you to configure and build POLARIS again).

The paper for this strategy uses an alpha to weight the state of charge (SOC) reward and a beta to weight the slack variable. The alpha in the JSON file is called “EV_LEVEL_UTILITY_INCREASE” and the beta is called “time_tradeoff”. To prevent repositioning of vehicles with relatively low SOC values, the parameter “EV_REPOSITIONING_MIN_SoC” sets the cutoff SOC value. Within the mass balance constraint, SAEVs with a SOC value below the “EV_SUPPLY_LEVEL_SOC” are not counted towards the zonal supply. To accomodate some queueing at charging stations, but not so much as to create traffic spillbacks, the parameter “EV_QUEUE_FACTOR” is used to scale parking spaces with the size of the station.

The joint_evcr strategy requires setting the following variables:

{
	"Joint EVCR Strategy" :{
	  "Repositioning Faculty": {
		"repositioning_flag": true,
		"demand_window_size": 1.0,
		"time_tradeoff": 500,
		"alpha": 1.0,
		"EV_REPOSITIONING_MIN_SoC": 45,
		"EV_LEVEL_UTILITY_INCREASE": 100.0,
		"EV_SUPPLY_LEVEL_SOC": 40,
		"EV_QUEUE_FACTOR": 0.30		
	  },
      "Charging Faculty": {
        "EV_CHARGE_IF_IDLE_FLAG": false,
        "EV_SERVE_IF_CHARGING_FLAG": false
      }
}

Day-Ahead Charging#

The day_ahead_charging strategy is a multi-stage charging and discharging strategy for SAEVs. This strategy finds the least cost hours to buy and sell electricity for the fleet and then conveys energy transactions into a new optimization-based dispatch strategy at discrete timesteps. This strategy uses the CPLEX solver (which may require you to configure and build POLARIS again).

The day_ahead_charging strategy requires setting the following variables (with changes made to prices_per_MWh, miles, and emission_damages_usd_per_MWh):

{
	"Required Strategy Parameters": {
		"MIN_EV_SoC": 15.0,
		"CUTOFF_BATTERY_LEVEL": 95.0,
		"SERVICE_FLAG": true,
		"SERVICE_CONVENIENCE_FLAG": true,
		"SERVICE_DISTANCE": 2.0,
		"repositioning_flag": true,
		"dump_solutions": false
    },
	"Charging Faculty": {
        "EV_CHARGE_IF_IDLE_FLAG": false,
        "EV_SERVE_IF_CHARGING_FLAG": false
    },
    "Fleet_Repositioning": {
		"Operator_1_TNC_OPERATOR_REPOSITIONING": false,
        "Operator_1_TNC_OPERATOR_REPOSITIONING_CHARGING": true,
        "zones_multiplier": 0.25,
        "demand_window_size": 1.0,
        "alpha": 1.0,
        "time_tradeoff": 2200,
        "EV_LEVEL_UTILITY_INCREASE": 85.0,
        "EV_SUPPLY_LEVEL_SOC": 40.0,
        "EV_QUEUE_FACTOR": 0.30
    },
    "Day_Ahead_Charging": {
        "plan_dayahead_charging": true,
		"total_charge_reduction_factor_AM": 0.20,
		"energy_sell_discount": 0.75,
		"demand_charge": 0.0,
        "cycle_cost": 0.025,
        "BEV_energy_efficiency": 0.279,
        "roundtrip_efficiency": 0.95,
        "min_fleet_SOC": 0.20,
        "max_fleet_SOC": 1.00,
		"medium_term_average_periods": 5,
		"medium_term_average_target": 0.60,
        "penalty_charge": 70,
        "penalty_discharge": 52.5,
		"target_midnight_soc": 0.90,
		"weight_target_soc": 2500000,
		"prices_per_MWh": [70,70,70,70,70,70,70,70,70,70,70,70,70,70,70,70,70,70,70,70,70,70,70,70],
		"emission_damages_usd_per_MWh": [67.16,79.56,80.67,77.94,77.57,70.46,63.29,61.42,70.97,56.45,56.78,49.16,43.39,41.68,45.89,48.57,52.7,49.81,50.62,54.31,50.77,41.07,40.15,56.18],
		"miles": [18066.735032, 14137.726564, 8601.972711, 12194.844088, 27154.670571, 58496.403231, 134083.717557, 278741.811107, 339517.408977, 271805.641915, 207598.491224, 226079.629124, 227718.980212, 215152.713144, 208484.549293, 242123.617472, 243882.458579, 218946.460946, 197338.135378, 135904.506715, 74675.783993, 48911.342576, 33814.023665, 14354.202931]
    }
}